WO2009056205A1

WO2009056205A1 - Configuration of a laser scanning microscope for raster image correlation spectroscopy measurement and method for conducting and evaluating such a measurement

Info

Publication number: WO2009056205A1
Application number: PCT/EP2008/008271
Authority: WO
Inventors: Stephan Wagner-Conrad; Frank Hecht; Klaus Weisshart; Ralf Netz
Original assignee: Carl Zeiss Microimaging Gmbh
Priority date: 2007-10-30
Filing date: 2008-09-30
Publication date: 2009-05-07
Also published as: US20100225910A1; US8120771B2; DE102007052551A1; DE102007052551B4

Abstract

1.1. The invention relates to a method for configuration of a laser scanning microscope for raster image correlation spectroscopy measurement and a method for conducting and evaluating such a measurement 2.1. The manual adjustment of the scanning parameters for a raster image correlation spectroscopy measurement (RICS) is expensive because the effects of an adjustment of a certain parameter are not obvious due to the complex interaction between the various parameters and also depend on the physical and technical properties of the microscope. Using an improved configuration method, mathematical transport models should be adaptable with a low incidence of errors to correlations determined by means of scanning fluorescence spectroscopy. Using an improved method for conducting or evaluating an RICS measurement, the amount of data to be stored should be reduced and it should be possible to detect RICS correlations with a higher statistical quality in a short time. 2.2. According to the invention, a best value is determined for a scanning parameter for a raster image correlation spectroscopy measurement and simulated to a sample for a later scanning process. In order to conduct or evaluate an RICS measurement, scan values are captured and a correlation is determined, only in a sample area inside which a pixel time (ΔP) changes along a harmonically controlled scanning axis (X) by less than or at the most by a predetermined or predeterminable value. 2.3. The invention is preferably used in laser scanning microscopes.

Description

Configuration of a laser scanning microscope for a

Rasterbildkorrelationsspektroskopiemessung and method for performing and evaluating such a measurement

The invention relates to a method for configuring a laser scanning microscope for a raster image correlation spectroscopy measurement and to methods for performing raster image correlation spectroscopy measurement and for evaluating samples of a raster image correlation spectroscopy measurement. Moreover, the invention relates to a correspondingly configured control unit, a correspondingly arranged laser scanning microscope and a correspondingly configured computer program.

To investigate variable substance concentrations in the microscopic size range, caused by diffusion and other transport processes in a sample, the so-called fluorescence correlation spectroscopy (FCS) can be used. Thus, physical and biological transport processes can be observed in a single or by a single volume with a diameter of about 200 nm. In order to be able to make statements about the transport processes in the sample, correlations of the fluorescence measurement data are determined and mathematical transport models are adapted to these correlations, for example by means of compensation calculations. For example, from the adapted models, diffusion constants can be determined. The transport models are typically mathematical functions whose parameters are adjusted.

Spatial resolution of microscopic transport events is provided by scanning fluorescence correlation spectroscopy (S-FCS), also known as

Image Correlation Spectroscopy (ICS). Here temporal orders of magnitude from seconds to minutes can be tracked. Two- or three-dimensional spatially resolved tracking within a cell or between membranes separated cells in the time domain between micro- and milliseconds is made possible by raster image correlation spectroscopy (RICS). Here, the sample is optically scanned in two or three dimensions in a grid-like manner For scanning correlation spectroscopy, a laser scanning microscope is expediently used. During the optical scanning movement of an RICS measurement, sampling samples ("sampling values") along a first scanning direction ("scan line") are electronically picked up and processed into pixel values at a typically constant sampling frequency , Each pixel value is determined from one or more samples. In this case, an interval within which a number of sampled values are assigned to a specific pixel or the period or frequency of such intervals is referred to as the pixel time. The scan along the first scan direction is repeatedly performed after the scanning beam has been displaced along a second scanning direction ("scan column") so as to record a sequence of pixel lines Scanning movement along the first scanning direction and constant sampling frequency, the time step size is varied in the form of the pixel time for a distortion-free recording with a constant spatial step size.

The errors in the adaptation of the mathematical transport models to the correlations determined on the basis of the scanning or pixel values depend directly on the suitability of the sampled values or the pixel values derived therefrom for such an evaluation. The suitability of the sample or pixel values, in turn, depends in a complex manner on the size of the sample area to be examined, the density of the fluorescence markers in the sample, contamination of the sample holder with fluorescence markers, the sampling frequencies and speeds, the optical properties of the microscope, the type of scanning movement , the illuminance and the sensitivity of the detectors. The concept of suitability includes statistical and systematic sources of error.

The invention is therefore based on the object of specifying methods and devices of the type mentioned above, with the help of which mathematical transport models can be adapted with low error to correlations determined by means of scanning fluorescence spectroscopy.

The object is achieved by methods having the features specified in claims 1, 16 and 20, and by arrangements having the features specified in claims 24, 25 and 26. Advantageous embodiments of the invention are specified in the subclaims.

According to the invention, for a method for configuring a laser scanning microscope for a raster image correlation spectroscopy measurement, it is provided that a best value for a scanning parameter is determined and predefined for a later scanning process on a sample. In particular, best values for a plurality of scanning parameters can be determined and specified.

For the purposes of the invention, scanning parameters ("scan parameters") are all those variables which can be set in an RICS measurement on the laser scanning microscope, ie the scanning movement ("scan movement") of a scanning device, also as a scanning device or as a scanner referred to, relative to the sample or the electronic recording of the light reflected or emitted by the sample influence.

A best value in the sense of the invention is a single value or even a range of values in which an evaluation of a RICS measurement with the least error or within an interval around the smallest error is possible. The evaluation may include, for example, the determination of pixel values from samples as well as auto and cross correlations and the subsequent adaptation by means of compensation calculation. Prior to evaluation, the samples or pixel values may be modified appropriately, for example by filtering. If there are several local error conditions, the best value can consist of several individual values or value ranges. The default can then be done for example in the form of a selection list.

Subsequently, instead of a sample value, a pixel value derived from one or more samples may always be used.

The manual adjustment of the sampling parameters for an RICS measurement is expensive, because the effects of setting a particular parameter by the complex interaction between the various parameters are not obvious and also depend on the physical-technical properties of the microscope. In particular, therefore, the expected suitability of a measurement for a single combination of settings of all parameters can only be roughly predicted. By the inventive automatic determination of a best value for one or preferably several such Parameter, the preparation and configuration of a RICS measurement can be greatly facilitated and accelerated. In particular, the physical-technical properties and settings of the microscope can be taken into account with high accuracy. As a result, mathematical transport models with a low error can be adapted to correlations determined from the sampled values.

Preferably, one or more of the following quantities are used as sample parameters: spatial step size, sample rate, pixel time, line time, sample rate. By scanning speed ("scan speed") is meant the spatial speed at which the scanning point or - when using illumination and detection slit diaphragms instead of pinhole diaphragms - the scanning line ("line scanning") move over the sample. The variables mentioned interact with one another, so that a manual configuration requires a great deal of knowledge about the mode of operation of the microscope and the scanning process and is therefore expensive. In particular, a constant scanning amplitude ("scan amplitude") is made possible by the use and optimization of these scanning parameters For this purpose, for example, according to the invention, the pixel and line times are set exclusively via the scanning speed, in particular taking into account the sample recording frequency.

In an advantageous embodiment, the best value of the spatial step size is determined based on a width of a point transfer function of the laser scanning microscope so that the spatial step size is significantly smaller than the width of the point transfer function (PSF). This represents an optical oversampling. In this way, the attenuation of the RICS correlations can be kept low, which is synonymous with a high statistical quality and thus a small error of subsequent functional adjustments. For example, the correlation for the case of free diffusion in three spatial dimensions results as

Therein, ξ, ψ pixel coordinates, fr the spatial step size of the scan, τ _P the pixel time, τ _L the line time, D the diffusion constant, ß ^ and ω _z the width of the PSF in the lateral and axial direction, N the number of fluorescent particles in the Confocal volume and γ a form factor, which takes into account the shape of the confocal volume. If the spatial step size ά small compared to the lateral width W ₀ of

PSF is set, the exponential factor ("scanning term") always remains in the order of 1, whereby a high suitability of the samples for the evaluation is possible.

Expediently, the best value of the spatial step size is determined on the basis of at least one objective parameter and a wavelength of light wavelength to be used during later scanning. As a result, a high accuracy of the RICS evaluation can be achieved.

In a particularly preferred embodiment, the best value is determined by carrying out several test scans of the sample to be scanned later, each with a different sampling value for the scanning parameter. This approach can be performed similarly to a trial-and-error strategy with little effort.

In this case, the best value can be determined by searching the minimum of the particular statistical error of a target variable, in particular by determining a minimum of an error of a variable characterizing a dynamics of a sample procedure from the sample values of the test samples. Preferably, to determine the target size and its error for each trial scan, a correlation is determined and a model function adapted thereto. The target variable can be an erroneous parameter of the model function or at least derived from such parameters.

Preferably, a diffusion constant is used as a variable characterizing the dynamics. Alternatively or additionally, the best value can be determined by determining a maximum of a width of a correlation of a test scan from the samples of the test scans. This can also be used with little effort as a measure of the statistical quality of the correlation.

Advantageously, in the experimental scans, the sample is bidirectionally scanned with the test value of the sampling parameter to be optimized different in the outward and in the backward direction. Thus, the number of trial scans can be reduced. The configuration can be done faster. Embodiments in which the scanning in the forward and backward direction takes place within the same scanning line are particularly advantageous. In this way, in one scan line, samples having two different pixel times can be acquired at the same line time without reducing the number of lines per direction. With only one sampling of the sample area to be scanned, three time parameters can already be checked. This not only reduces the illumination time of the sample, but also enables the configuration with high speed and high accuracy. Such an embodiment of the method can also be used with line-shaped illumination and scanning by means of slit diaphragms. This type of capture captures multiple pixels in the same scan column simultaneously. The pixel time has the same value for all pixels of the scan column. By a bidirectional scanning along a scan line so two pixel times can be tested in a trial scan.

Preferably, a simulation of a transport process in the sample and / or a virtual test scan are performed to determine the best value. For example, the virtual trial scans may be performed by modeling the sample to be examined and resulting samples or by changing real measured samples to avoid sample damage due to too frequent or too long exposure. In particular, correlation data can be determined only mathematically by simulation using a model, for example a model of dynamic particle interactions. Thus, the exposure of the sample to the actual measurement process can be limited using the previously virtually determined best value.

In a preferred embodiment of the invention, the sample is actually over-sampled electronically and at least one virtual test scan based on the oversampling is performed by subsequently mathematically varying the number of samples taken for the virtual test scan. This provides a quick way to tentatively determine the sampling parameters pixel time and line time with different values and to determine the value with the best suitability without having to perform a real sampling with these pixel and / or line times. In particular, different values of the sampling parameter "spatial step size" can be set experimentally for this purpose. In order to vary the number of samples, advantageously several of the samples may be summarized or omitted. This can be done with high speed and high accuracy.

In an advantageous embodiment, the sample area to be scanned is determined on the basis of geometric specifications of a user. The sample area to be scanned may consist of one or more, each in themselves related areas or there are treated several separate sample areas in parallel. For each area or each sample area then results own best values for the one or more sampling parameters. The geometric specifications are preferably made on the basis of an overview image of the sample graphically by selecting regions of interest of the sample (ROI). Both in the configuration and in the RICS measurement then preferably only the predetermined areas / areas are scanned. The RICS evaluation is performed by correlation analysis, preferably including cross-correlation, and also preferably only in the areas of interest.

In such embodiments, the best value is preferably determined exclusively for and exclusively on the basis of samples or derived pixel values from the sample area to be scanned. This reduces the duration of the configuration process.

In special embodiments of the invention, the best value is output as a suggestion and a confirmation of a user for the optimized value is determined before a later scanning process. By proposing the best value or a range for the sampling parameter, the user has the option of checking and modifying the best value before the actual measurement. Conveniently, the user interface offers the possibility to set the proposed best value by an operation for the subsequent scanning or select from the proposed area.

According to the invention, for a method for evaluating samples of a raster image correlation spectroscopy measurement of a sample it is advantageously provided that a correlation is determined exclusively in a sample area within which a pixel time along a harmonically controlled scanning axis is less than or at most by a predetermined or predeterminable value, in particular by 10%, changes. Thus, RICS correlations with high statistical quality can be determined in a short time. Accordingly, for a particular method of performing a raster image correlation spectroscopy measurement on a sample, samples are taken only in a sample area within which a pixel time along a harmonic driven scan axis is less than or at most a predetermined or specifiable value, in particular, 10%. , changes. This reduces the amount of data to be stored. Samples outside the range of near-constant pixel time, that is, samples whose associated pixels have pixel time varying by more than 10%, are not stored.

In a particular embodiment, for a high suitability of the sampling parameter, sampling values can be determined exclusively in a plurality of sampling regions, within which the pixel time changes along the harmonic-controlled sampling axis by less than or at most by a predetermined or predefinable value. Accordingly, for a method for evaluating samples, a

Rasterbildkorrelationsspektroskopiemessung a sample advantageously provided that separate correlations are determined in several scanning areas within which changes the pixel time along the harmonic driven scanning each by less than or at most by a predetermined or predetermined value. The maximum values for the deviations in the different scanning areas may differ or be uniform, for example not more than 10% in each area. The total correlation G then results as a superposition of the individual correlations G _{1 of} the individual scanning regions, for example to:

Where ξ, ψ are pixel coordinates, <5r is the spatial step size during scanning, τ _{P is} the pixel time, τ _{L is} the line time, D is the diffusion constant,

ω _{z is} the width of the PSF in the lateral or axial direction and N _{1 is} the number of fluorescent particles each contained in the confocal volume. According to the invention, the sample can preferably be oversampled optically and / or electronically for a good statistical quality of the correlations.

Advantageously, the samples are recorded by means of a laser scanning microscope. Appropriately, a correlation is determined from samples of a raster image correlation spectroscopy measurement by means of a control unit of a laser scanning microscope.

In this case, the number of sampled values can preferably be changed by combining or omitting the correlation in order to simulate different sampling parameters. Thus, the sampling parameter can be varied after the actual data acquisition, considering the resulting suitability of the thus obtainable samples for the correlation and model analysis, in particular in one or more virtual test scans.

The invention also encompasses a control unit for a laser scanning microscope, which is set up programmatically for carrying out a method according to the invention, a laser scanning microscope with such a control unit and a computer program which is set up to carry out a method according to the invention.

The invention will be explained in more detail by means of exemplary embodiments.

In the drawings show:

1 a laser scanning microscope,

2 shows diagrams of the scanning movements and the pixel time,

3 shows the restriction of the data acquisition and correlation range,

4 shows the division of the sample area to be scanned into sections with a nearly constant pixel time,

5 shows the scanning with different scanning speed in the forward and reverse directions, 6 shows the variation of the number of samples,

Fig. 7 shows the selection of areas of interest and

8 is a flow chart of a configuration, data retrieval and evaluation method.

In all drawings, like parts have like reference numerals.

FIG. 1 shows, by way of example, a confocal laser scanning microscope, which is particularly suitable for carrying out the method according to the invention. Therein a laser module 1 is provided, which is equipped with lasers 2, 3 and 4 for generating laser light of the wavelengths 633 nm, 543 nm and 488 nm for the visible range. The radiation emanating from the lasers 2, 3 and 4 is coupled via several beam combiner 5, an AOTF 6 and an optical fiber 7 into a scanning device 8, which is equipped with a unit 9 which deflects the beam in the coordinates X and Y. The scanning movement along the scanning axis X can take place in the form of a harmonic oscillation or, for example, in the form of a triad progression. In a second laser module 10, a UV laser is provided, whose light is coupled via an AOTF I l and an optical fiber 12 in the scanning device 8.

In both beam paths, the optical fibers 7 and 12 are followed by respective collimating optics 13 whose distances from the respective fiber end are variable and which are coupled for this purpose with a controllable adjusting device (not shown in the drawing).

From the beam-deflecting device 9, the laser radiation is coupled through a scanning lens 14 in the beam path of the microscope 15 shown simplified and here directed to a sample 16 containing a fluorescent dye or to which such a dye has been applied. On the way to the sample, the laser radiation passes through a tube lens 17, a beam splitter 18 and the microscope objective 19. The light reflected and / or emitted by the respectively applied location of the sample passes through the microscope objective 19 back to the beam deflecting device 9, then passes through a beam splitter 20 and, with the aid of the imaging optics 21 after splitting into a plurality of detection channels 22 on detectors 23 in the form of

For the purpose of splitting into the individual detection channels 22, the light is directed by way of example by a deflecting prism 24 onto dichroic beam splitters 25. In each detection channel 22, the photomultiplier tube (PMT) is directed to one of the detection channels 22 Both adjustable in the direction and perpendicular to the radiation direction and changeable in their diameters Pinholes 26 and emission filter 27 are provided.

The outputs of the detectors 23 lead to the signal inputs of an evaluation circuit 28, which in turn is connected to a control unit 29 in order to integrate and evaluate the signals of the detectors 23 pixel by pixel. The outputs of the control unit 29 are connected to the signal inputs of the laser modules 1 and 10, with signal inputs of the adjusting devices for influencing the position of optical elements or assemblies, such as the position of the collimating optics 13 and the pinholes 26, with the scanning device 8 and with the evaluation circuit 28 connected (not shown in detail). The control unit 29 is also connected to a display (not shown) and controls (not shown) for data entry and modification.

By way of example, the laser radiation coupled into the scanning device 8 is branched by a beam splitter 30, one of the branches for monitoring the laser radiation being directed to an optoelectronic receiver 31, the plurality of filter filters arranged on filter wheels and interchangeable by rotation of the filter wheels line filter 32 and also against each other replaceable neutral filter 33 are arranged upstream. The output of the receiver 31 is also located at a signal input of the evaluation circuit 28. The filter wheels on which the line filter 32 and the neutral filter 33 are arranged are coupled to adjusting devices whose control inputs are connected to signal outputs of the control unit 29 (not shown in the drawing).

Of course, the method according to the invention can also be implemented with laser scanning microscopes of significantly simpler construction. In FIG. 2, for the case of a sinusoidal harmonic motion, one of the deflecting mirrors of the beam deflecting unit 9 in subfigure 2A is the deflection angle φ of the relevant deflecting mirror, in subfigure 2B the deflection X of the scanning beam on the specimen 16 and in subfigure 2C the location-dependent pixel time τ _P shown. The pixel time Tp is the duration over which a signal from a photomultiplier 23 is integrated into a sample for a respective pixel and averaged. Due to the harmonic motion, the pixel times Xp at the outer edges of the scannable area are significantly longer than in the center. Analytically, the deflection X of the scanning beam in the sample plane is given as: x - f tan [fesin (ö) t)] - fb sin (<»0 _» where t is the time, b the

Amplitude and ω indicates the rotational frequency and the approximation for small deflection angle is applicable. The local step size ΔX = δr is preferably kept constant in order to record undistorted images. This results in a changing time step size Δt or pixel time Tp: At = τ _P = Ax / r ^" -'lb. ^' - ^] 1 ω ^~

COS (BJ /)

3 shows the local limitation of data sampling in an RICS measurement with a harmonic scanning movement of the light beam along the scanning axis X and a linear scanning movement along the scanning axis Y. FIG. 3A shows the restriction in a deflection time. Diagram, sub-figure 3B locally in an image (XY-diagram) of an exemplary sample. In order to be able to apply model functions based on the assumption of a constant pixel time Tp to the determined pixel values with a low error, the RICS evaluation takes place according to the invention only in a restricted middle X-region in which the pixel time Tp is less than 10 % changes. In alternative embodiments, even in the regions with larger deviations, the data may be analyzed with model functions adapted to the changing pixel time Tp. In a specific further embodiment of the data acquisition (not shown), for example, the evaluation circuit 28 generates samples only while the scanning beam is in the middle X-range. This reduces the amount of data and the amount of data to be stored.

The data acquisition of Fig. 4 is not limited in such a way. Rather, the sample is subdivided into several regions i = 0... 3, which are assigned to different individual correlations G i. Within each of the areas, the pixel time Tp ,, does not change by more than 10% each. It is thus approximately a piecewise linearity of the scanning accepted. For each area, the average pixel time Xp _is stored for RICS analysis.

Before the sample 16 is scanned for an actual RICS evaluation, best values for individually selectable scanning parameters can be determined by means of the control unit 29 at the request of a user / user and proposed on the display. This is possible regardless of the type of scanning movement (harmonic, triangular, etc.). The user / user then has the option of using the operating elements to adopt the proposed scanning parameters for a subsequent scanning process or to modify them beforehand. Alternatively, the sampling parameters can preferably be compulsorily optimized after the start of an actual RICS measurement as part of the measurement process before the actual sampling, without informing the user. This also allows users / users who do not have a detailed knowledge of the optical and electronic scanning process to determine model parameters of low-error transport processes using RIC S measurements and evaluations. With the same effect, it is conceivable that particularly qualified operating personnel undertakes an optimization of selected or all scanning parameters according to the invention outside normal operation and specifies the best values thus found or a value of a best-value interval thus found for subsequent RICS measurements and evaluations in regular operation that are performed by less qualified users. Best values can be determined, for example, for the interacting scanning parameters spatial step size δr and pixel time Xp.

A best value for a scanning parameter can be determined purely mathematically on the one hand, and on the other hand through a number of experimental test scans with varying scanning parameters. For example, a best value for the spatial step size δr can be determined purely mathematically from the objective parameters and a wavelength of light to be used during later scanning. For example, the best pixel time Xp and the best line time X _L can be determined purely by real trial scans, each with a different value for times Xp, X _L. Mixed forms are also conceivable, for example the proportional calculation of the best value for the spatial step size δr and the proportional experimental determination for the pixel time Xp, while the line time X _{L is} not optimized, but left to the user for free adjustment. Instead of real experimental scans (test measurements) in combination with a variation of at least one scanning parameter, the best value can be determined by purely computational simulations on the basis of dynamic models. This is useful, for example, to minimize the light energy deposited in the sample when the sample is phototoxic and there is enough information about the dynamic transport process for modeling.

It is also possible to perform an artificial variation of at least one scanning parameter based on one or more real experimental scans. This can be considered as a virtual trial scan. For example, a real sample parameter such as the pixel time Tp may be modified by combining or omitting different numbers of samples of a real trial sample.

For both the purely computational simulation of virtual test scans and virtual trial scans based on one or more real test measurements, the controller (not shown) may provide a user interface. The determined best values are then given automatically or on demand for a subsequent RICS measurement and evaluation.

From multiple trial scans, whether real or virtual, a best score can be obtained by taking from the samples of the trial scans the RICS correlations and, after fitting a model function by a compensation calculation, a minimum of one sample dynamics error characterizing target size, for example, a diffusion constant, is determined. This can be done, for example, in a per se known interval bisecting method until a minimum of the target size error is found.

5 shows the solution according to the invention for accelerating the configuration method in the case of a scan with an approximately triangular course, but the solution can also be used for harmonic motion. The sampling of the sample in a real trial scan is bidirectional. In this case, from the control unit 29 in the forward scan a first scan speed V _{X LR of} the beam deflecting unit 9 from left to right and in the back scan a second scan speed Vχ _{, RL} of beam deflecting unit 9 set from right to left. This results in a different pixel time T _P , L- _R in the forward scanning direction than in the reverse scanning direction with T _P , R. _L - The data acquisition takes place only in two areas, for example, in which the pixel times Tp _1R-L and Xp _{1R-L are} approximately constant. In an alternative embodiment (not shown), the data recording can take place over the entire sampling amplitude, wherein for the evaluation only samples from the two regions with approximately constant pixel time Tp, _L - _R or Tp _1R-L are used. In both alternatives, the data of the two scanning directions are processed separately to pixel values (image data) and analyzed separately with respect to the optimization of the scanning parameters. By varying the pixel time Tp between a plurality of complete test scans of the sample area to be scanned, after correlation formation, adaptation of the model function and error determination of the target parameter, the best value can be determined on the basis of the pixel time Tp at which the error of the target parameter is minimal.

In Fig. 6, an embodiment is shown in which, instead of changing the scanning speed Vx, the pixel time Tp is varied by reducing the number of samples taken in an electronic oversampling, more specifically the averaged raw data of the evaluation circuit 28. This can be done by combining or omitting samples, in particular with a constant sampling frequency of the evaluation circuit 28. As a result, the number of pixels available for correlations is reduced, requiring a corresponding change in the effective spatial increment δr of the virtual trial scans. By way of example, the spatial step size δr can be increased as the pixel time Tp increases, for example when the number of samples per pixel is increased.

The sample acquisition frequency is, for example, constantly 40 MHz. The scanning speed V _X ^ .R _{/ R} - _L ) is set by the control unit 29 so as to effect electronic oversampling. By means of different variants of the reduction of the number, virtual scanning tests with different pixel times Tp, _k can be carried out in order to determine the best value. For example, a virtual pixel time Tp _{> 2} can be determined virtually from a pixel time Tp, ι by averaging two adjacent sample values. The pixel time Tp _{> 2} is then twice as large as Tp _, i. The reduction of the number of samples can be advantageously combined with optical Oversampling can be used because the exponential factor can be kept almost constant.

In alternative embodiments (not shown), the sample acquisition frequency can also be varied as a sampling parameter.

FIG. 7 shows several regions of interest RO1 that have been selected graphically by a user on the basis of an overview image of the sample. The configuration recording and evaluation methods according to the invention can be carried out separately and independently of the other regions of interest in each region of interest. Cross-correlations between the areas are also possible in the RIC S evaluation.

FIG. 8 schematically summarizes the steps of a method according to the invention, which includes both the configuration of the laser scanning microscope and the data acquisition and evaluation for the determination of difusion constants D ₁ in various regions of interest i, in a flow chart. The fast scan axis X is controlled harmonically by way of example, the slow scan axis Y linearly.

LIST OF REFERENCE NUMBERS

1 first laser module

2,3,4 laser

5 beam combiner

6 AOTF

7 optical fiber

8 Scan setup

9 Beam deflecting unit

10 second laser module

1 1 AOTF

12 optical fiber

13 collimation optics

14 scan lens

15 microscopes

16 sample

17 tube lens

18 beam splitters

19 Mikroskopobj ekti v

20 beam splitters

21 imaging optics

22 detection channel

23 secondary electron multipliers

24 deflection prism

25 beam splitters

26 pinholes

27 emission filter

28 evaluation circuit

29 control unit

30 beam splitter

31 Optoelectronic receiver

32 line filters

33 neutral filter

Vx _1L-R Hin scanning speed vχ, R- _L Reverse scanning speed φ deflection angle

X, Y deflection, scanning axes

Δt time increment δr Spatial increment

Tp pixel time

ROI Interesting area i Number of the area of interest

G, Gj correlation

F model function

D diffusion constant

Claims

claims

A method of configuring a laser scanning microscope for a raster image correlation spectroscopy measurement, comprising the steps of:

Determining a best value for a sampling parameter and

- Specifying the best value for a later scan on a sample (16).

2. The method of claim 1, wherein one or more of the following quantities is used as the sampling parameter: spatial step size, sampling rate (vχ), ₅ pixel time (τp), line time, sampling rate.

3. The method of claim 2, wherein the best value of the spatial step size is determined based on a width of a point transfer function of the laser scanning microscope so that the spatial step size is significantly smaller than the width of the point transfer function.

4. The method of claim 2 or 3, wherein the best value of the spatial step size is determined based on at least one lens parameter and a light wavelength to be used in later scanning.

5. The method according to any one of the preceding claims, wherein the best value is determined by a plurality of experimental scans of the sample later to be scanned (16) are performed, each with a different experimental value for the scanning parameter.

6. The method of claim 5, wherein the best value is determined by determining from the samples of the trial scans a minimum of an error of a variable characterizing a dynamics of a sample event.

The method of claim 6, wherein a diffusion constant is used as a variable characterizing the dynamics.

8. The method according to claim 5, wherein the best value is determined by determining a maximum of a width of a correlation (G) of a test scan from the samples of the test scans.

9. The method according to any one of claims 5 to 8, wherein in the test scans the sample is bi-directionally scanned with in the outward and return direction different experimental value of the parameter to be optimized.

10. The method of claim 9, wherein the scanning in the forward and backward direction is within the same scan line.

11. The method of claim 10, wherein to vary the number of samples, a plurality of the samples are combined or omitted.

12. The method according to any one of the preceding claims, wherein for determining the best value, a simulation of a transport process in the sample (16) and / or a virtual trial scan are performed.

The method of claim 12, wherein the sample (16) is real over-sampled electronically and at least one virtual trial scan based on the oversampling is performed by computationally varying the number of real samples for the virtual trial scan.

14. The method according to any one of the preceding claims, wherein the sample area to be scanned is determined based on geometric specifications of a user.

15. The method according to any one of the preceding claims, wherein the best value is determined exclusively for and exclusively on the basis of samples from the sample area to be scanned.

16. The method according to any one of the preceding claims, wherein the best value is output as a suggestion and before a later sampling an acknowledgment of a user for the optimized value is determined.

17. A method for performing a raster image correlation spectroscopy measurement on a sample (16), wherein samples are taken only in a sample area within which a pixel time (Tp) along a harmonic driven scan axis (X) by less than or at most by a predetermined or predetermined value , in particular by 10%, changes.

18. The method of claim 17, wherein samples are determined exclusively in a plurality of scan areas within which the pixel time (τp) along the harmonic scan axis (X) changes by less than or at most by a predetermined or specifiable value.

19. The method according to claim 17 or 18, wherein the sample (16) is oversampled optically and / or electronically.

20. The method according to any one of claims 17 to 19, wherein the samples are recorded by means of a laser scanning microscope.

21. Method for Evaluating Samples of a

Rasterbildkorrelationsspektroskopiemessung a sample (16), wherein a correlation (G) is determined exclusively in a sample area within which a pixel time (τp) along a harmonic driven scanning axis (X) by less than or at most by a predetermined or predetermined value, in particular 10%, changes.

22. The method according to claim 21, wherein separate correlations (G i) are determined in a plurality of scanning regions, within which the pixel time (τ p) along the harmonic driven scanning axis (X) changes by less than or at most by a predetermined or predefinable value.

23. The method of claim 21 or 22, wherein for the determination of the correlation (G), the number of samples is changed by combining or omitting.

24. The method according to any one of claims 21 to 23, wherein the correlation (G) by means of a control unit (29) of a laser scanning microscope are determined from samples of a raster image correlation spectroscopy measurement.

25. Control unit (29) for a laser scanning microscope, which is set up by the program to carry out a method according to one of claims 1 to 23.

26. Laser scanning microscope with a control unit (29) according to claim 25.

27. Computer program, set up for carrying out a method according to one of claims 1 to 24.